9556

• UCREFRP Catalog Number: 9556
• Author: Bestgen, K. R., K. A. Zelasko and C. T. Wilcox.
• Title: Non-native fish removal in the Green River, Lodore and Whirlpool canyons, 2002-2006, and fish community response to altered flow and temperature regimes, and non-native fish expansion.
• USFW Year: 2007.
• USFW - Doc Type: 115,
• Copyright Material: NO

Collection ofsmall-bodied fish.-Low-velocity channel margins in Browns Park, Lodore
and Whirlpool canyons, and Island-Rainbow Park were sampled with seines (1.3 and 4.6 m
length, 1.6 and 4.7-mm mesh). Seine sampling was conducted in spring, summer, and autumn in
each year from 2002 to 2004, and during summer and autumn in 2005 and 2006. Habitat types
sampled were mostly backwaters, but channel margin eddies, low velocity runs and pools, and a
few riffles were also sampled. More effort was expended in larger habitat areas and less in
smaller ones so that the proportion of the habitat sampled was approximately equal across
different-sized areas. Riffles were occasionally "kick-seined", whereby substrate was vigorously
disturbed and dislodged fish were captured in a stationary downstream seine. Some vouchers
were preserved in 10% formalin and identified at the Larval Fish Laboratory, Colorado State
University. We scrutinized all young chubs captured and preserved in 2002 to 2006 by counting
dorsal and anal fin rays, and characterizing shape characteristics of the snout and mouth (mouth
terminal in roundtail chub Gila robusta, slightly subtenminal in humpback chub), body depth, fin
lengths, and the line of the angle of the anal fin base relative to the upper lobe of the caudal fin,
which assisted with identifications (Muth 1990, Douglas et al. 1989, Snyder et al. 2006).
Drift net samples collected in the Green River just upstream of the Yampa River in 2002
to 2006 were used to describe species composition and abundance patterns of fish larvae
transported downstream and to determine if upstream spawning of Colorado pikeminnow had
occurred. Samples were collected daily in the morning (beginning ca. 0630 to 0900 h) with
conical drift nets (0.15 mZ mouth diameter, 4 m long, 560 µm mesh) set near shore in water 30 to
50-cm deep. Three nets were set on each sampling occasion for up to 2 h, but sampling ceased if
debris load exceeded 3.8 L/sample. Water depth at which a white object disappeared from sight
was recorded as a measure of water turbidity. General Oceanics flow meters (model 2030)
suspended in each net mouth recorded water velocity. Samples were fixed immediately in 100%
ethanol and fish were removed from debris within 4 h and preserved in 100% ethanol. Ethanol
was used to ensure that fish otoliths or other tissues were useful for later analysis, if desired.
Collection oflarge-bodied fish.-Most collections of large-bodied fishes were by
electrofishing and (mostly in Whirlpool Canyon) trammel nets; angling was used on a limited
9